hkz and bubbliiiing
9dbb4f88d4
Update 7b && Support low vram inference ( #196 )
...
---------
Co-authored-by: bubbliiiing <3323290568@qq.com >
2025-02-12 10:06:07 +08:00
hkz and bubbliiiing
f74a6cb427
Support TeaCache ( #180 )
...
* Support TeaCache
* Update app
* Update teacache
---------
Co-authored-by: bubbliiiing <3323290568@qq.com >
2025-01-23 15:52:48 +08:00
Bubbliiiing and hkunzhe
9f34f7c8c5
Update to V5.1 ( #179 )
...
* Update Flow
* Update Flow
* Update Flow
* add image recaptioning
* Fix bug in t2v
* update train_reward_lora.py
* update reward training
* Update V5.1 and mix multi text_encoders to one pipeline
* Update V5.1 training Code
* Update ComfyUI
* Update Comment
* Delete files
* update reward training
* Update Readme
* fix extract frames in compute_semantic_consistency
* Update Readme && Remove to in prediction
* Update Demo
* Update Readme
* Update ui
* support vae gradient checkpointing in reward training
* Update Training Readme
---------
Co-authored-by: hkunzhe <huangkunzhe.hkz@alibaba-inc.com >
2025-01-22 15:30:47 +08:00
Bubbliiiing
45c298af71
Update Readme && Update lora cache && Lora merge with CUDA ( #173 )
...
* Update Readme
* Fix bug in training
* Update Readme
* Update Readme && Update lora cache
* Update Readme
* Update error info in sampler
* Fix Memory bug in 12b
2024-12-26 15:41:21 +08:00
Bubbliiiing
94f420c257
Low ram support. ( #154 )
...
* Fix bug in install comfyui
* support model load in 30GB RAM
* Update requirement
* Update requirement
* Update VAE outlier_penalty_loss
* Fix bug in ui
* Fix bug in ui
* Fix bug in ui
2024-11-20 15:56:28 +08:00
Bubbliiiing
9da6b79ece
Update Prompt && Update Readme && Fix control bug ( #131 )
...
* Update Readme
* Update prompt tips
* Fix bug in ui
* Fix control bug and Update readme
* Update prompt again
2024-11-11 13:48:53 +08:00
Bubbliiiing
62de94e2f1
Update V5 ( #128 )
...
Update V5
2024-11-08 19:46:50 +08:00
e4f1a4fe97
Update V4 version ( #92 )
...
* update train_lora && update deepspeed && update training with max token length
* fix bug in train.py
* fix bug in training_with_video_token_length
* update v2v && update v2v api
* add rope2d embedding precomputation; move text encoder to dataloader to reduce gpu memory consumpution
* add cuda multi-stream to speedup vae encode
* update new vae && new comfyui
* fix some bug in training code
* Add lcm lora (#89 )
Co-authored-by: xuanyuan.lb <xuanyuan.lb@alibaba-inc.com >
* Update Training Code and fix bug in low vram mode
* fix bug in low vram mode
* update report
* update cfg
* actual text clip
---------
Co-authored-by: mengli.cml <mengli.cml@alibaba-inc.com >
Co-authored-by: liubo0902 <38622806+liubo0902@users.noreply.github.com >
Co-authored-by: xuanyuan.lb <xuanyuan.lb@alibaba-inc.com >
2024-08-19 11:22:17 +08:00
bubbliiiing
039d67acf1
support float16 && add reference && fix bug in training
2024-07-24 13:19:03 +08:00
bubbliiiing
2eef3f9f78
update v4
2024-07-18 16:33:56 +08:00
Bubbliiiing and chenyunkuo.cyk
f9eeabe231
Updated to v3 version, supports image generated videos, with a maximum support of 960x960x144 video generation. ( #40 )
...
* update v3
* Fix frame start_idx bug, see issue #41 .
* update readme and fix bug in training
* update requirements
* update ui
* fix bug in inpaint
* update new ui
* fix bug in auto resize
* fix bug in auto resize
* fix bug in modelscope and eas
* update low gpu memory mode
---------
Co-authored-by: chenyunkuo.cyk <chenyunkuo.cyk@alibaba-inc.com >
2024-07-05 20:27:24 +08:00
